Sukhpinder Singh

Posted on Mar 26 • Updated on Mar 27 • Originally published at singhsukhpinder.Medium

Day 1 — Azure Open AI Challenge

#openai #chatgpt #ai #programming

Learnings include creating Azure Open AI services, deploying GPT models, interacting with GPT in Playground, and optimizing model performance with parameters.

Introduction

The Microsoft team has recently launched an Open AI challenge, in which a developer can learn how to build Azure AI solutions and apps.

Prerequisite

Experience working with Azure and Azure portals.
An understanding of generative AI.

Getting Started

Step 1: Navigate to the Azure Portal

Search for Azure Open AI and fill out the following details

Step 2: Choose Network

Step 3: Add Tags

For this exercise tag names are not required. But in a production environment, it should be added as it's a best practice.

Step 4: Review & Create

Step 5: Move to Open AI Studio

On the left side open Models > Select “gpt-35-turbo” > Deploy as highlighted below

The below model list will be populated based on the access provided by the Microsoft team.

Azure OpenAI includes:

GPT-4 models: Latest generation for natural language and code completions from prompts.
GPT 3.5 models: Generate language and code completions, with turbo versions for chat interactions.
Embedding models: Convert text to numeric vectors, useful for language analytics.
DALL-E models (preview): Generate images from natural language prompts, automatically available without explicit deployment in Azure OpenAI Studio.

Step 6: Open Playground

Step 7: Try Prompts

Input the following prompt in the chat to see if the service is working or not. It may take 5–10 minutes, please be patient.

hi can you tell me about viral kohli

Please find below the response to the above prompt.

An alternative approach

The whole setup can also be done via the CLI using the following command

az cognitiveservices account deployment create \
   -g OAIResourceGroup \
   -n MyOpenAIResource \
   --deployment-name MyModel \
   --model-name gpt-35-turbo \
   --model-version "0301"  \
   --model-format OpenAI \
   --sku-name "Standard" \
   --sku-capacity 1

I haven’t tried the CLI approach but just wanted to mention the other alternative.

Playground Parameters

Play with playground parameters to change the performance of the model.

In this module, we learnt

How to create an Azure Open AI Service.
How to deploy GPT models.
How to converse with GPT in the Playground.
Different approaches to deploying an Open AI service on Azure.
Playground parameters are available to improve the performance of the model.

C# Programming🚀

Thank you for being a part of the C# community! Before you leave:

If you’ve made it this far, please show your appreciation with a clap and follow the author! 👏️️

Visit our other platforms: GitHub | Instagram | Tiktok | Quora | Daily.dev

DEV Community